The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes data from external Google Spreadsheets which serves as an ingestion point for untrusted content.
Ingestion point: The read_sheet tool in handler.ts fetches cell values and formats them into the agent's context.
Boundary markers: The formatted output string (tab-separated rows) lacks explicit delimiters or instructions telling the agent to ignore any commands found within the sheet data.
Capability inventory: The skill possesses write capabilities (write_sheet, append_to_sheet), which could be abused if an attacker places instructions in a spreadsheet that the agent then follows to modify other data or spreadsheets.
Sanitization: There is no evidence of filtering or escaping logic to prevent the AI from interpreting cell content as instructions.

google-sheets